Muster travel speed has been drastically reduced. You will no longer be able to amass your full army nearly instantly at any rally point. Equally, dismissing your armies will be quite a risk in a war, since they will take much longer to travel back to their homes. This is actually a rather nasty nerf for a large empire, as it can take quite a while to actually get a full army assembled. This has no effects on MAA units.

Second, levies do not regenerate with nearly the speed they once did. You can and will be worn down if you fight wars constantly with heavy losses. Again, a nerf for a larger empire, which due to it's size may find itself in many smaller wars. MAA regeneration is also reduced, but only slightly, with the cost adjusted to match. Losing masses of men is no longer a trivial matter.

Finally, army movement itself is slowed slightly, even when raised. Armies will not zip around the map, this makes positioning and supply more important to consider, and again limits the ability of a single army to respond to multiple smaller armies. Sea travel is also quite a bit slower, though days without attrition has been adjusted to allow the same total range, just without the insanely rapid speeds. Smaller, more compact realms will see rather large benefits from these three changes, and equally larger realms will feel their weight as the burden it should be.

::::

Fertility is something which is...much more fragile. You will, on average see fertility reduce more when you are older, and base childbirth is reduced as well. You will see far fewer children on average, meaning that bonuses to fertility have more inherent benefit. This is, of course, not entirely realistic, but it is better than having a third of children dying around age 1-2 and lagging the game with useless data (And the result is the same). Also that would be a bit depressing.

Children, and women, are more likely to die in childbirth, or have other outcomes which are not ideal, this is especially the case when the mother is in poor health.

While less children can make succession easier with partition laws, this may also harm the player, since the AI is far less able and willing to murder, disinherit or cheese their way to a single heir. AI realms will be a bit less prone to fracturing, at least from succession. There is also the very real danger of not having ENOUGH children, meaning that ill timed deaths can result in real problems. Your idiot fifth cousin is now your heir, have a great day!

::::

Religious tenets have been rebalanced, with the goal of making more tenets viable or interesting. There should be, ultimately, no 'bad' tenets that are never worth taking in any situation. This will improve difficulty in the game, as the AI generally won't make their own religion, and thus the player will min-max and many default religions had weak tenets by default. Improving these 'bad' tenets helps this situation, and thus base religions will be stronger by comparison to their prior selves, or against the player min-max religion.

A few religious choices were changed to fit history, mostly in cost, but in a few cases availability. Sacred childbirth, for example, was not an option for Christian faiths... despite Catholicism's huge fixation on Mary... seems like it should at least be possible, since naked Christians who live in forests are an option. A lot of minor cost adjustments, as well, just to smooth out some odd situations with religions.

::::

The economy has been changed, with basic buildings providing less income and levies, but the effects of county development has been doubled. This makes defending your realm far more important, since rich developed lands can be twice as valuable, even with the exact same buildings. Terrain also plays more of a part, affecting building costs and development growth more than vanilla.

::::

Adjusted genetic inheritance to be less deterministic. This will mean that concentrated breeding programs will be less effective, and the trait will randomly inherit a bit more often. This should help the AI for obvious reasons, and make looking for marriage partners more interesting since more active traits will be wandering around the world. Also inbreeding will be less super effective, always a good thing.

::::

There is an additional difficulty option added, called roleplay difficulty. It is now considered the default difficulty, and applies a 25% decrease to stress losses, only for the player. It also increases all tyranny gains by 25%. This is meant to keep the player more bounded in, while still providing them a level playing field with the AI. You may find yourself having to care more about your character's traits, and perhaps a little less able to rule recklessly without irritating your vassals.

Rolehard has been added as further increase to roleplay enforcement, with a 50% to both values.

::::

Vassals have been given quite the rework, starting with the most basic change that if allied to their liege, they can be called into defensive wars by their liege. This means realms may have a defensive advantage, however all is not perfect. Vassals are no longer quite so willing to make alliances, and even with an alliance they are less willing to enter a internal war. All alliance calls have been a bit more complex and less automatic, meaning that every alliance is not a blanket guarantee of support, especially when the ally likes the attacker and defender both. Expect more negative opinions in general, since there has been significant bloat of positive opinion modifiers with T&T and Artifacts. You will need to consider more carefully the personalities of your vassals, since if they clash with you those modifiers are more potent. Equally vassals are far more unhappy with higher crown authority, meaning that keeping crown authority low may improve realm stability. And of course, making alliances within your realm can protect you, since allies will never join factions against you, but beware, since you cannot imprison or revoke the titles of the vassals who have pledged to be your allies.

::::

Sway and many other schemes are now more uncertain, even in ideal circumstances. Sway especially has been heavily nerfed, only providing up to +50 increased opinion at maximum. There will be characters who hate past the ability of sway to fix even with time. Sway also takes into account some of the more serious actions the player may have taken in the past, applying success penalties. If you murdered someone's child, swaying them at all will be an uphill battle for obvious reasons. Many hostile schemes now take far more account of intrigue scores of the target, meaning that high intrigue can provide extensive protection. At the same time, the handcuffs have been unlocked for the AI, they will murder players if they hate them and see an opportunity. Higher ranks also receive a bit more hostile scheme resistance, so have fun trying to murder an emperor without agents or a severe intrigue advantage.

::::

There are also an increasing number of general changes involving:

-> several changes which make religious unity less simple to achieve, especially among vassals.
-> Cultural difference opinion penalties increased
-> Dread changes, closing loopholes which allow dread to be gained without some cost
-> Mercenaries are noticeably cheaper at base price, encouraging their use in more cases
-> Random lifestyle xp from events now more significant.
-> Some balance work on dynastic legacies.

Most of these changes again are problems for empires more then smaller powers, reflecting the real difficulties of maintaining a large empire that might contain other cultures and religions. This may also benefit religions and strategies which sacrifice pure power builds for those with more tolerance. Which is a weak playstyle already, so a good balance outcome.